<![CDATA[Enhancing the performance of academic staff is a key factor in maintaining institutional productivity and service quality in higher education. This study aims to develop a data-driven hybrid model capable of enhancing performance management effectiveness through the integration of predictive intelligence and evidence-based managerial recommendations. This model combines K-Means Clustering, Data Envelopment Analysis (DEA), Exploratory Factor Analysis (EFA), and Random Forest to analyze digital attendance data, service satisfaction surveys, and performance records from 2022 to 2024. This research was conducted at the Faculty of Teacher Training and Education, Sebelas Maret University (FKIP UNS) as a representative case study. The test results show that the predictive model achieved 92 percent accuracy and an F1-score of 0.90 in classifying low performance risk. A strong negative correlation was found between attendance tardiness and service satisfaction levels. DEA analysis identified 32 percent inefficiency in resource utilization, while EFA revealed three dominant latent factors: compliance with SOPs (0.82), academic productivity (0.89), and psychosocial well-being (0.93). Intervention cluster management (SOP training and workload reduction) resulted in a 28 percent increase in SOP compliance. These findings indicate that the integration of hybrid machine learning with efficiency and factor analysis can be an effective framework for data-driven improvement in academic staff performance.]]>
Copyrights © 2025
